In 2006, Medicare introduced Part D outpatient prescription drug benefits, which could have significant implications for patients requiring chronic and high cost drug therapy, such as antipsychotics. Medicare provides formulary protection for antipsychotics;however, these drugs are subject to potentially substantial and complex cost-sharing requirements included in many Part D plans, such as spending-based coverage gaps. In general, coverage limits have been associated with decreases in necessary drug use, increases in adverse clinical events, and increases in medical costs that offset pharmacy cost savings;however, there is a dearth of evidence on the effects of Part D benefit policies on patients prescribed antipsychotics. This study will examine the impact of Part D coverage on 1) drug use and adherence;2) adverse clinical events, including emergency department visits, hospitalizations and death;and 3) medical spending, including out-of-pocket drug spending, and pharmacy, outpatient, inpatient, and total medical costs. The study population includes beneficiaries receiving antipsychotics (n=43,217 in 2007) who are enrolled in Medicare Advantage Prescription Drug plans offered by two large Part D sponsors, 2004-2010. It includes plans that vary in cost-sharing levels during the gap, beneficiaries receiving low income subsidies (37% of the sample), and beneficiaries with employer- sponsored insurance, whose benefits do not include a gap or change over time to control for secular trends. We will compare beneficiaries with higher versus lower Part D cost-sharing (e.g., basic gap vs. no gap), as well as examine the impact of low income cost-sharing subsidies. The study period includes two years pre-Part D, 2004-2005, and five years post-Part D, 2006-2010. We will use fixed-effects (conditional) estimation methods to examine changes in annual outcomes associated with Part D cost-sharing levels, as well as within-year, temporal changes associated with the coverage gap (e.g., before and after the reaching the gap). Defining the study cohort based on drug class is consistent with the level at which Medicare has made national formulary policy;however, because we expect the impact of cost-sharing to vary by the clinical appropriateness of antipsychotic use and the availability of treatment alternatives, we will assess variations by psychiatric diagnosis. We will adjust for relevant demographic, clinical, and plan characteristics using Medicare Prescription Drug Event files, as well as plan membership, clinical encounter, and pharmacy data. This study provides a unique opportunity to link Part D drug utilization with clinical event and medical spending data for a large, well-defined population of Medicare beneficiaries prescribed antipsychotic drug therapy. The study findings may inform improvements to the program, such as the need for specific cost-sharing exemptions (e.g., value-based benefit designs), completely or partially filling in the coverage gap, or targeted subsidy expansions, to reduce unintended consequences and improve the value of Part D for beneficiaries using antipsychotics. PUBLIC HEALTH RELEVANCE: Nearly all Medicare Part D drug plans involve substantial and complex patient cost-sharing, including a coverage gap or donut hole. Patients prescribed antipsychotic drug therapy for mental illness may be particularly vulnerable to limited drug coverage;however, little is known about their experiences under Part D. This study will examine the effects of Part D cost-sharing, overall and by psychiatric diagnosis, on drug use, adverse clinical events, and medical spending using comprehensive automated data from before and after the introduction of Part D for a well-defined population of beneficiaries prescribed antipsychotics.